(2) Discussion on the method to estimate the time serial changes of tidal front 
There should be many cases where tidal front location data of NOAA may not be available 
due to weather, etc. It is considered necessary to study the method to estimate the time serial 
changes of tidal front location not only based on the NOAA data which are obtained separately 
but also combining it with the simulation models based on oceanic and meteorological data. 
(3) Discussion on the enhancement of tidal front extraction method 
It is considered necessary to study some expert extraction method combining the pattern 
extraction from the smoothed data and rule base based one the tidal front properties. 
References 
(1) Akaike, H: Likelihood and Bayes procedures, Bayesian Statistics (Bernardo, J. M., De 
Groot, M. H., Lindiey, D.U. and Smith, A.F.M. eds.) University Press, Valencia, 
pp.143-166,1980. 
(2) Tsunakawa, H.: Baysian approach to smoothing palaeomagnetic data using ABIC, 
Geophys. J. Int., 108,pp801 - 811 , 1992. 
(3) Yasuaki Murata, Harufumi Noro: Optimization of calculation method of lithosphere surface 
layer density based on ABIC minimization estimation method, Physical Survey Vol. 46 No. 
2, 120 - 127, 1993. 
(4) Genshiro Kitagawa: Time serial analysis programing, Iwanami-shoten 
(5) (Foundation) Japan Survey Association: lustrated Remote Sensing, Japan Remote Sensing 
Research Associationed. 
52